[00:06] robert_ancell: i pushed a branch to wip/williamhua/ubuntu-sso [00:06] robert_ancell: it basically just hooks into ubuntu-sso-client's dbus service to get the oauth tokens [00:15] robert_ancell: oh. btw, i've only tested the token retrieval... i didn't want to test sending the actual review to the production server so i inserted a 'return TRUE' locally [00:32] attente, awesome! [00:36] attente, you need to replace all your spaces with tabs. Silly tabs. [00:37] attente, oh, my bad [00:37] they already are. I was just confused by the output of git diff [00:40] I'll take tabs over spaces any day. :) [01:12] attente, do I need to install anything? It didn't prompt me for anything [03:49] robert_ancell: it shouldn't need anything besides ubuntu-sso-client i think [03:50] but even that i think should be optional since it queries it over d-bus [03:50] i mean from a "won't crash if it isn't installed" perspective [03:50] Yeah, it didn't crash, but it didn't post the review either [03:51] did it at least get the tokens for you? [03:51] i didn't test the posting the review yet because i didn't want to test aginst the production server [03:53] attente, I don't think so [03:53] attente, there were some build issues, but that branch should now be merged and in the PPA [03:54] robert_ancell: gdbus call --session --dest com.ubuntu.sso --object-path /com/ubuntu/sso/credentials --method com.ubuntu.sso.CredentialsManagement.find_credentials_sync 'Ubuntu One' '{}' [03:55] attente, that seems to get them [03:55] can you add a printf in the plugin to see if it also still gets them? [03:58] attente, I'm about to sign off - I'll check next week [03:59] robert_ancell: sure, i'll test it again tomorrow [03:59] have a good w.e. [03:59] attente, feel free to update the PPA [03:59] sure, not sure if i have rights though [04:00] actually, one last question: there isn't a development server to test on is there? [04:03] attente, I think there is https://reviews.staging.ubuntu.com [04:04] I never tried it though [04:04] attente, one of the issues is you have a 30min window to re-review and then you can't change them. This was changed for phone apps, I have a MP to the reviews server to change the behaviour there too. [04:06] does that also mean that reviews aren't published during that window? that might make testing a bit slow... [04:07] attente, I think they are - you just get 30min to change your mind [04:08] bye all [07:44] good morning [08:26] happy Friday! [08:28] happy friday larsu [08:28] morning hikiko! [08:28] and pitti :) [08:29] good morning :) [08:30] hey hikiko and larsu, guten Morgen! [08:30] hallo pitti! === Guest83051 is now known as fredp === fredp is now known as Guest84471 [08:57] morning all [08:57] stupid Google now asks me which account I would like to use every time I open my calendar. [08:58] I've only got one account [08:58] sigh [08:58] true story [09:04] heeeeeeeeeey [09:05] good morning all! [09:06] maybe it counts both the gmail and the canonical account? [09:07] hi alexarnaud [09:07] and willcooke [09:12] hey Laney alexarnaud [09:13] hi willcooke, happy friday [09:13] \o/ [09:14] I have https://www.google.com/calendar/b/1/render?pli=1#main_7 for calendar [09:14] morning Laney and willcooke! [09:14] always goes to the right account [09:14] morning larsu [09:14] I think "1" is the account id [09:14] ah, nice one - thanks Laney [09:14] heeeeeeeeeeey larsu! [09:15] good morning Laney! [09:15] hey willcooke [09:15] hi hi pitti [09:15] what up pitti [09:15] so I solved my array issues last night [09:15] took me faaaar to long [09:15] the array was indeed being initialised with zeros [09:15] but [09:15] I'd #define ARRAY_LENGTH 4 [09:16] thinking 0,1,2,3,4 = 5 [09:16] ah, statically sized? [09:16] hah [09:16] willcooke: #define? I thought you write C++? [09:16] Is this hybrid Arduino C/C++ [09:16] probably not the best thing to learn on [09:16] but flashing leds make me hapy [09:16] happy [09:16] willcooke: for a statically sized array, it's much safer to do something like char* myarray[] = {"first", "second", "third", NULL}; [09:17] I know how big the array can be, but not what will be in it at compile time [09:17] Laney: did you say you fixed the issue with the text in the scope for online and offline? [09:17] why? [09:17] no [09:17] hikiko: did you reveive my mail ? [09:17] alexarnaud, let me check I wasn't here yesterday [09:18] pitti, so I was running off the end of the array and that was making for for loop count to 46 instead of 4. [09:18] Laney: ah so it was a fix for something else then that just happened to coinside with me confirming the issue for willcooke then \o/ [09:18] I fix EVERYTHING all the time! [09:19] willcooke: if you know the array size, you can iterate over it using that, and not depend on a NULL terminator [09:19] pitti, I was just using a simple for (int x=0; x <=sizeof(array); x++) [09:19] note the " <= " :) [09:19] willcooke: < [09:19] hikiko: OK, take your time. We haven't had problem with our mail serveur it's why I ask to you ^^. [09:20] willcooke: ah :) [09:20] willcooke: but that shoudl be off-by-one, not off-by-42 [09:20] mitya57: was it you that fixed ap-gtk? [09:20] yeah, that's what totally confused me. How running off the end of the array made it count to 46. Probably an MCU thing. Maybe I just got unlucky and poke the wrong byte in memory [09:21] it was good fun though. [09:21] alexarnaud, I can't find it [09:21] Sat here until 10pm last night. Shouting at the thing [09:23] willcooke: shouting with gdb usually helps :) [09:23] :D [09:23] The debugging on Arduino is, as far as I can tell, limited to Serial.println [09:24] \o/ [09:24] willcooke: expletives OTOH .. one should not anthropomorphize computers too much .. they don't like that! [09:24] lol [09:24] wfm ;) [09:24] physical violence works better [09:24] willcooke: ah, yay, good old printf debugging [09:24] * Laney remembers our boiler mallet back in student days [09:24] good times [09:25] "screw this, I use python" :) [09:25] Laney: "boiler" + "mallet"? that awfully sounds like "oops, I flooded the appartment" [09:25] :D [09:26] it soudned like an explosion waiting to happen anyway :) [09:26] hitting it actually improved matters [09:26] *nnng* I don't want to know. I don't, no, no [09:26] in other news: I just submitted my first autopkgtest retry request over the web UI [09:28] oh! [09:28] web ui? [09:28] http://10.0.3.78:5000/?release=trusty&package=coreutils&arch=amd64&trigger=glib2.0/2.40.0-2 [09:28] ha [09:28] (don't bother, this is a local LXC) [09:29] Laney: adding a "retry this test" button to excuses.html which devs can use [09:29] now, back to cranking test coverage back up to 100% [09:29] yeah I know the feature [09:29] not kept up with what you achieved at the sprint [09:30] Laney: I'll blog about that today, just wanted to finish this up [09:30] * Laney eyes glib/ppc64el [09:35] hikiko: I re-send it to you now [09:37] mitya57: oh, it was autopilot-legacy, disregard [09:38] oh alexarnaud I just saw both they were in spam folder and I hadn't sync it let me read it [09:40] ok alexarnaud so the problem is that you change something and the gsettings are not stored right? [09:41] or they are stored but you see no difference? [09:42] hikiko: there are stored in Unity profile instead of mate-accessibility profile [09:42] alexarnaud, 1st of all are you running unity+mate or ubuntu+mate? [09:42] sorry [09:42] and in CCSM the profile swiching produce no effect [09:42] unity+compiz [09:43] or mate+compiz [09:43] Debian+Mate+Compiz [09:43] the mate accessibility is only useful when you are on mate desktop [09:43] oh so you set something for mate [09:43] and is set for unity? [09:44] It is set on the general section of the config file [09:44] do you want the configuration file maybe ? [09:44] is it the xml file? [09:45] I use two files : [09:45] 1) /etc/compiz*/config (I don't remember the exact name) [09:45] and I use an override file with our (hypra.fr) parameters [09:46] oh maybe compiz doesn't look at this file :) [09:46] the first define gsettig backend and mate-accessibility as default profile [09:46] hikiko: oh, really? [09:46] so the problem is [09:46] that you dont see the custom settings you wrote in the file [09:46] right? [09:47] The config file produce effect I want [09:47] but gsettings DB doesn't match it [09:47] so all configuration are stored in the wrong profile [09:48] settings are stored in profile Unity instead of mate-accessibility. CSSM shows the default profile is mate-accessibility but no configuration in it. [09:48] ok probably dconf and compiz look at different files [09:49] alexarnaud, because I don't know how you configured compiz and dconf (paths etc) what I would do to be sure [09:49] hikiko: is profile switching work on your computer when you used gsettings? [09:49] is to add the configuration options in the xml files of compiz [09:49] what is the xml file of compiz? [09:49] give me a second (I don't have mate I think) let me check [09:50] The gsettings schema maybe? [09:50] alexarnaud, which settings did you change? [09:50] what I would do is to add my files to compiz before installation and then install it [09:50] hikiko: lets me alse a minute to give you all configuration and defails [09:50] sure [09:54] hikiko: the override file http://paste.ubuntu.com/14596794/ [09:55] hikiko: the config file http://paste.ubuntu.com/14575900/ [09:55] (in /etc/compiz*) [09:56] /etc/compizconfig maybe? [09:57] alexarnaud, and in dconf-editor [09:58] which are the settings you look at? [09:59] atm I dont have mate I only see org.profiles.unity [10:00] .plugins and list [10:01] alexarnaud, if you run gsettings set org.compiz.pluginname:variable value do you still have the same problem? [10:10] hikiko: I will make you a dump of gsettings [10:11] sorry someone come to meet me [10:12] heh. This is funny. Starting synapse from "inside" desktop leads to a crash ( sigsegv in libgdk-3) everytime. Starting it in a tty (ctrl+alt+f4) makes it work. [10:18] okay, so I found the culprit. unsetting GTK_IM_MODULE=xim makes it work "inside" again. Is this a bug or not? [10:18] or is my install/conffiles incorrect somewhere? [10:22] this happens in 15.10 btw ^ [10:28] hikiko: http://paste.ubuntu.com/14596877/ [10:28] hikiko: this is the dump of dconf org.compiz [10:29] I hope you have all informations you want [10:29] :) [10:30] alexarnaud, I am not sure what's wrong it might be a bug I will look at it but maybe not today because I have to finish something else first before I install mate to experiment [10:34] mitya57: also it looks like the new kwallet backend for python-keyring is chosen too aggressively? [10:34] it's breaking the ubuntu-sso-client build at least it seems [10:35] hikiko: OK [10:37] Laney, one more item to my TODO list :) But a wild guess is that a new upstream release may fix it. [10:38] (With upstream's "release several times a day" principle I'm quite lagging behind) [10:38] mitya57: funny that your name keeps coming up this time :) [10:38] mitya57: I'm trying to install a config file in the package build to select a different backend [10:38] maybe that works [10:39] If you can test the latest release that will be also nice :) [10:39] too much red on glib/excuses atm [10:39] Laney, do you have a build log link btw? [10:39] maybe soon [10:39] can boucne you one [10:40] done [10:41] second one is after adding xvfb-run -a dbus-run-session [10:42] Laney, yes, 7.3 should fix it [10:43] Maybe I'll even upload it today if I have time [10:43] cool [10:44] hikiko: hi [10:45] hi ksamak [10:45] sorry about the incomplete questions. the actual question is rather like so [10:45] we're using mate with compiz, with gsettings backend, and an .override file [10:46] at install with our packages [10:46] although the gsettings are set up fine, we have trouble identifying if profiles are worth someting with gsettings in compiz. [10:47] so, is compiz only recognizing one sole profile when on gsettings backend? [10:47] or can we hope to have several profiles configured differently? [10:47] note: we are using ccsm to configure/change profiles, maybe it's not up to date [10:48] we have doubts because changing profiles in ccsm doesn't change the actual final settings. [10:49] and dconf-editor show our modifs hierarchically under the "unity" compiz profile [10:49] hikiko: [10:51] ksamak, as far as I know (I am not too familiar with the gsettings either) you can have many profiles but one profile per backend [10:51] For instance, if you are using the GSettings backend then any new profile you create will be a GSettings profile. If you switch to a different backend then your current profile will not work and you will automatically switch to a profile available for that backend. [10:51] just found that on the arch wiki here: https://wiki.archlinux.org/index.php/Compiz_configuration#Profiles [10:52] so I think what you suspect is correct [10:52] yeah but we tried to implement a second gsettings profile. [10:52] and the settings don't actually change between profiles. [10:52] ok. thx [10:52] everytime you create a new profile the settings go to unity? [10:53] mmm andyrock Trevinho could you help? ^^ [10:54] well, it seems that whatever profile is active (all gsettings), doesn't change the actual settings. [10:55] so what we override works, but we can't switch to any other. so i was wondering whether out override is correct [10:55] when you say doesn't change you mean that you see no changes in compiz or that you cant see them in dconf? [10:56] dconf displays settings under the unity profile. compiz behaves with overriden profiles, not changing when switching profiles [10:56] if you restart compiz? [10:57] reminder: andyrock Trevinho http://paste.ubuntu.com/14596794/ [10:58] hikiko: shouldn't we have a profile specified in our override by the way? [10:58] that's what i was wondering [10:58] cause this looks like global settings [10:58] as for profile-less settings like for mate or so [11:02] yes they look global and to be honest I never switched profiles to change the settings I have to look at it ksamak I don't know because I am not familiar with that part of code, it looks more reasonable that the setting should be per profile we might have a bug I'll ask Trevinho too (he is at a conference this week) [11:03] also I'll install mate to reproduce the problem, there might be something with the settings [11:05] i don't think this is linked with mate, but ok thanks for your interest [11:05] i guess people don't use ccsm so much? [11:05] yes probably I can check with a random new profile [11:06] ksamak, usually people change the settings from ccsm for the profile that is already selected (either default or unity) [11:06] yeah no-one ever changes profiles, right lol [11:07] we're the only ones thinking that that might be nice XD [11:07] we actually want to setup different profiles for different types of handicaps. [11:07] then we should definitely look at this because it sounds like a bug [11:08] for more flexibility with handicapped people [11:08] yes, I understand [11:09] well I ll look at it next week I believe it's a compiz or ccsm bug [11:09] you should be able to have settings per profile [11:26] thx [12:12] morning [12:12] hikiko: ksamak do you still need help? [12:13] morning andyrock [12:14] hi andyrock :) [12:15] yes the question is: how could someone in compiz have different gsettings for different profiles [12:15] it seems that we have a bug and when you select a profile in ccsm [12:15] and you set some gsettings they go global [12:15] they apply to all profiles [12:16] ksamak, had this configuration file: https://paste.ubuntu.com/14596794/ [12:16] sorry: this: https://paste.ubuntu.com/14596794/ [12:17] he uses the existing unity and mate profiles [12:17] but when he sets gsettings for mate [12:18] they go to the unity section in dconf-editor [12:18] and they are global [12:18] I am not very familiar with that part of compiz :/ [12:19] if you could help it would have been nice otherwise I'll look at this problem next week [12:19] sorry bbiab [12:51] i'm not familiar with this part of compiz neither [12:52] hikiko: we can wait for Trevinho or I can try to debug it if it's urgent [12:53] when is Trevinho back? [12:54] lemme check the spreadsheet.. [12:54] best case scenario is that marco knows a trick to do it without ccsm :) [12:55] Flight is on Monday, so maaaaybe Tuesday PM, most likely weds [12:56] it's actually no sleeping [12:56] I was speeking with him few minutes ago on the phone [12:56] but it's better to don't bother him at this time :D [12:56] he needs to sleep too [12:58] :) [13:01] e ok let's get a look on monday then and when marco is back he helps us I don't know if it's urgent but if it's a bug I guess is important (no?) I don't know === hikiko is now known as hikiko|ln [13:13] mm ksamak [13:13] i have a more recent ccsm and dconf [13:13] I created a foobar profile [13:14] (new one) [13:14] added some settings [13:14] restarted dconf-editor [13:14] and they appear there [13:14] ksamak, Compiz 0.9.12.2 that's my version [13:14] dconf-editor 3.18.2 [13:15] but I set the settings through ccsm [13:15] not manually [13:15] so I believe that it's just something in the settings [13:16] let me try to import your settings now [13:21] ksamak, when I select the ccsm profile and set the settings for it in ccsm they are applied to the profile only [13:21] when I import your .profile [13:21] they are globally applied [13:22] bbiab [13:26] ksamak, i think the syntax is not correct [13:27] ksamak try something like that: https://paste.ubuntu.com/14597543/ imported when your custom profile is selected and check dconf editor again I think it will work [13:27] it seems to work for me [13:27] basically try it with 2 settings only because that's huge :) [13:28] bbiab lunch break :) [13:48] re [13:50] hikiko|ln: ok, thanks for taking the time to test, it's kind of what we needed! [13:51] although your file is the same syntax as ours, right? [13:51] we can't really rely on that for an installation from .deb package though. [13:51] if you're right [13:51] i'll try and think again about that [13:52] no the difference is the [] [13:53] basically ksamak I exported your settings from 1 profile [13:53] oh, ok. [13:54] ksamak, when Trevinho is back he will help more [13:54] I am just experimenting I am not familiar with gsettings [13:54] indeed [13:54] there's a difference indeed. [13:54] so this makes it relative to a profile. [13:55] which is good. now we need to try and find a way to specify the profile applied. [13:55] there are some profile.ini files in my system [13:55] unity.ini [13:55] for example [13:55] but I don't have much time to look at this today [13:56] next week :) [13:56] I think they are relevant [13:56] if you want to get a look yourself [13:57] yeah these are the ini backend files. [13:58] we tested that, it works. but we need to migrate to gsettings. [13:58] hikiko|ln: thanks again for all your help [13:58] np :) [13:58] hope it helps [14:01] already does. === hikiko|ln is now known as hikiko [14:04] oh, its Friday! Lets continue the discussion on C++. [14:05] So [14:05] http://45.media.tumblr.com/e545c444e8e28687ce9363e84017e7d5/tumblr_nz3134g9dT1rn7bzro1_500.gif might be a starting point. [14:05] If I want to pass a 2 dimensional array as a pointer to a function.... [14:05] Sweet5hark1, love that gif :) [14:08] haha nice one [14:32] hikiko: thanks again [14:33] I've read the discusession [14:33] It seems that maybe I need to make another test also === daniel__ is now known as DSMcGuire === DSMcGuire is now known as DS-McGuire [15:36] Laney: hey, do you know if there are any docs for the software center rest api? [16:03] attente: short answer is no [16:03] attente: long answer is which API do you mean? [16:03] (that's not much longer!) [16:04] which ever one we need for gnome-software to submit reviews [16:04] i'm trying to add support for that there, but can't figure out why i keep getting a 401 error [16:05] ah, sorry, I thought robert_ancell had already done that though [16:05] but if I were to pick someone it would be m v o [16:05] is someone writing a new server with new api or something? [16:06] I thought he was just interfacing with the same thing software-center uses currently [16:07] yeah, i was just hoping there'd be an easier way than reverse engineering the old code [16:08] I bet not, but check with reverse(ovm) to be sure [16:08] lol [16:08] is that part not done in robert's branch(es) already? [16:09] it was missing the ubuntu sso which i added yesterday [16:09] ah [16:09] but i guess it wasn't testable because of the lack of sso at the time [16:09] yeah sorry, I've not worked on the client at all [16:09] just the server bit [16:10] if you're after another task then you could see why the metadata doesn't load up properly (email from me and matthias to ubuntu-desktop last week) [16:10] otherwise will look at that first half of next week [16:10] sure [16:11] is that server bit you have include a staging server for testing? [16:12] oh. just read it... [16:14] mvo: hey, do you know if there are any docs for the software center rest api? [16:14] yeah but not integrated [16:15] so you get to use that shitty script I wrote to wangle the files into place [16:25] It's friday :)! If you want to try your linux history knowledge you can deal with this quizz : http://fossforce.com/2016/01/how-well-do-you-know-your-linux-history/ [16:29] 15 / 18 - 2 of which were silly mistakes [16:45] willcooke: it's the same for me [16:45] I don't know anything about crises about legal rights [17:40] http://cdimage.ubuntu.com/ubuntu/daily-live/pending/ [17:40] haz images [17:52] Laney, hey! [17:53] hi seb128! [17:53] good stuff, thanks Laney [17:53] hey seb128 [17:53] what is up? [17:54] hey willcooke [17:54] smelling like w.e for you guys? ;-) [17:55] :D [17:55] willcooke, just had a chat with alecu, +1 for the "keep click but drop the packagekit interface" [17:55] seb128, awesome, thanks! Good news indeed [17:55] Laney, ^^ :D [17:56] cool [18:01] happy weekend! [18:02] bleh [18:02] can someone retry autopilot and autopilot-legacy once sphinx gets published please? [18:02] byeee [18:02] Laney, enjoy! [18:02] Laney, retry what? builds? [18:42] kenvandine, thanks for the reply on that bluetooth bug [18:44] go technologie, it's 2016 and you could expect that things get better [18:44] not that new protocol suck so much that you can't even know what type of device you are listing [18:45] yeah [18:46] case #1 is the pressing issue we have now [18:46] but... i'm more interested in case #2 [18:46] i've had to add several device types to our filter over the past 6 months or so [18:47] seb128, not sure about iOS, but android seems to let you pair anything too [18:47] well, android let you do things with devices [18:47] like send fles over obex [18:47] or use mouse pointers [18:48] which we don't [18:48] yeah, hopefully we will too :) [18:48] I find it a bit confusing that you can't pair a device that then does nothing [18:48] you mean can? [18:48] it feels like we have bugs [18:48] yes, sorry [18:48] i felt that way too :) [18:48] but simon convinced me i was wrong :) [18:49] hopefully someday we'll have uses for all of these :) [18:49] we can't have a dev convincing every user that they are wrong to feel like they do [18:49] well, we were always going to whitelist things that have an use [18:50] anyway if design is happy about it [18:50] let's see what's the user feedback [19:15] night all. Have a great time a SCALE, and safe travels for those of you on the way to FOSDEM and related activities. [19:15] Everyone else, happy weekend! [22:34] seb128: around? [22:34] Sweet5hark1, yes [22:35] seb128: heya, on a sprint or something? [22:35] Sweet5hark1, at UbuCon yes ;-) [22:36] seb128: anyway, 5.1.0~rc2 is build and in the ppa -- if there is no big complaints in the next days, i'd suggest to sponsor that into xenial. [22:36] seb128: ah, cool, have fun! [22:36] Sweet5hark1, sounds good to me, let me know if you need sponsoring [22:36] I'm travelling on monday and should be back on tuesday [22:37] seb128: rationale for going with rc2 already instead of final: winter releases are always a bit more tight and this release seems less risky than other major releases. [22:38] I'm all for not being overconservative and land the rc [22:40] seb128: this release also still has libmwaw libw{d,g,s} and lpolve bundled, but Istill think we should upload it as-is, we can than still fix that afterwards without doing everything in one step. [22:41] yeah, tuesday/wednesday sounds good. Ill be travelling to FOSDEM on Friday early morning. [22:43] seb128: alright, tuning out for the weekend. have fun at UbuCon! [22:52] Sweet5hark1, thanks, have a good w.e!